Sibley v. Sprint Nextel Corp.

315 F.R.D. 642, 2016 U.S. Dist. LEXIS 121499, 2016 WL 4679274

• Court: District Court, D. Kansas
• Decided: July 29, 2016
• Docket: Case No. 08-CV-2063-KHV
• Status: Published

Opinion

REPORT REGARDING MOTION TO DECERTIFY

David R. Cohen, Special Master

This Court appointed the undersigned as Special Master and, among other things, requested “a recommended ruling on the pending motion to decertify the class.” Appointment Order at 5 (docket no. 532). For the reasons stated below, the Special Master now recommends the Court DENY Sprint’s renewed motion to decertify (docket no. 619).

I. General Background.

Defendants Sprint Nextel Corporation and Sprint/United Management Company (collectively, “Sprint”) are in the business of providing wireless telecommunications products and services. In August of 2005, Sprint acquired Nextel, a smaller competitor. After the merger occurred, Sprint had problems marrying the two companies’ computer systems; these problems resulted in inaccurate calculations of amounts owed to commissioned sales representatives. Accordingly, in February of 2008, nine Sprint employees who worked in retail sales filed suit against Sprint. The thrust of their complaint was that Sprint routinely failed to pay them commissions they had earned and were owed. The nine plaintiffs currently assert claims for; (1) violation of the Kansas Wage Payment Act (“KWPA”); and (2) breach of contract.¹ The plaintiffs allege they suffered damages, in the form of improperly withheld commission payments, averaging between $100 - $500 or more per month. First Amended Complaint at 6 ¶ 23 (docket no. 8).

In addition to filing suit on their own behalf, the nine plaintiffs also sought to pursue identical claims on behalf of a nationwide class of Sprint employees who were similarly-situated. Accordingly, the plaintiffs filed a motion for class certification pursuant to Fed. R. Civ. P. 23 (docket no. 36). In November of 2008, the Court granted this motion and certified a class under Rule 23(b)(3). Specifically, the Court certified a class composed of “[a]U persons nationwide who worked for Defendants’ retail stores since their merger with Nextel [in August of 2005] ... whose compensation was based in full or in part on commissions.” Certification Order at 2 (docket no. 99).²

In its Certification Order, the Court concluded the defined class met each of the four requirements of Rule 23(a) — that is, numer-osity, commonality, typicality, and adequacy. Id. at 12-18. Further, the Court concluded the defined class also met both requirements of Rule 23(b)(3) — that is, predominance and superiority. Id. at 19. The Court noted, however, that, depending on how the case developed, it might be “later persuaded that a class action is not the most efficient form of litigating this controversy, [in which ease] it may decertify the class.” Id. at 22; see also id. at 10 (observing that courts “have broad discretion to later redefine (or even decertify) the class if necessary”). With the Court’s 2008 class certification ruling in hand, the parties proceeded with discovery and motion practice.

Splint and plaintiffs each hired experts to examine Sprint’s computerized commissions system and determine whether Sprint had accurately calculated the commissions it owed its retail sales representatives. The amount of work required of the experts to undertake this determination was massive. That is because, during the class period: (1) Sprint employed over 30,000 class members; (2) these class members engaged in a total of over 350 million potentially-commissionable sales transactions, which were documented by over 6 billion computerized records; (3) Sprint’s compensation scheme changed several times, creating over a dozen variations on how commissions were calculated; (4) each one of these compensation schemes was complicated, involving intricate assessments of sales and also different commission measures for different product categories; (5) the information necessary for the experts to calculate and reconcile historic commissions came from numerous sources — for example, the experts had to match information contained in (a) retail sales databases, (b) customer billing databases, (c) databases defining which products and services were commissionable, and (d) payroll databases, among others; (6) the relevant data was generated by and flowed through many different computer programs, which were not necessarily designed to “talk to each other;” and (7) as Sprint and Nextel merged them operations, Sprint changed the computer programs it used to calculate employee compensation.

Four years after the class was certified, the experts submitted reports setting out their conclusions. Of course, rebuttal reports followed, criticizing the opposing experts’ conclusions. Given the scope of the task, it was unsurprising that each side’s “first tries” at calculating the correct commission amounts contained errors or oversights— thus, each expert conceded that some of the opposing experts’ criticisms were valid. Accordingly, each expert submitted amendments to their reports — which, in turn, required amendments to Daubert motions. Even after several corrective amendments, however, the conclusions the experts reached were dramatically different. Although both sides’ experts agreed Sprint’s actual compensation payments to class members had rarely been accurate, plaintiffs’ two-expert team, known as Balance Engines (“BE”), concluded Sprint had underpaid the class by a total of $171.0 million, while Sprint’s expert, Dr. Janet Thornton, concluded Sprint had overpaid the class by a total of $72.9 million. See BE’s Third Supplemental Report at 5, ¶ 20 (Dee. 20, 2013) (docket no. 618-20); Thornton’s Affidavit at 30, ¶ 62 (Dec. 20, 2012) (docket no. 507-2).

Because the subject matter of the experts’ reports was so “highly technical and complex,” and because the Court was concerned “a jury will find the experts’ opinions so opaque that the jury will be forced simply to guess” who was correct, the Court appointed the undersigned to recommend rulings on pending Daubert and decertification motions. Appointment Order at 2 (docket no. 532). Later, the Court also appointed a Technical Advisor, Dr. Chen Song, “to help the Special Master and the Court understand how each expert: (1) translated the terms of Sprint’s sales commission plans into computer code, (2) interpreted, integrated, and cross-referenced Sprint’s numerous databases, in order to obtain a final result; and (3) used and created his or her particular methodology.” Order at 3 (docket no. 559).³

After the Court entered its Appointment Orders, the Special Master and Technical Advisor met with the parties several times. At these meetings, “the parties’ experts [offered additional criticisms of] the other side’s methodology and conclusions,” and Dr. Song also pointed out possible weaknesses her own reviews had revealed. Amended Scheduling Order at 1 (docket no. 579). Both sides’ experts admitted that even their “third tries” at calculating accurate commissions could still use refinement and improvement. Accordingly, “the parties, the Special Master, and Dr. Song ... all agreed that the parties’ experts should have one more opportunity to supplement their expert reports and rebuttal reports. The parties further agreed that, after filing these amended, supplemental expert reports, they should have one more opportunity to amend their briefs related to the pending Daubert motions, summary judgment motions, and the motion for class decertification.” Id. at 2. To make things simpler procedurally, the Court simply denied all pending motions without prejudice and directed the parties to submit comprehensive and unitary omnibus “renewed” motions, without incoiporating by reference earlier-filed documents. Id. at 4 n. 2; see id. at 3 (noting that, because the parties had “amended many of the pending motions and reports numerous times,” the Court would otherwise have to “review numerous separate documents, which refer to each other willy-nilly”).

Accordingly, in late 2015 and early 2016, the parties filed new expert reports, Daubert motions, and motions for summary judgment and decertification.⁴ Thus, it is Sprint’s renewed motion for decertification (docket no. 619) that is the subject of this Report.

II. Sprint’s Contracts with its Retail Salespeople.

To understand Sprint’s decertification arguments, it helps to understand the contracts at the heart of this case. During the four-year class period, Sprint used 13 different compensation plans. Specifically, Sprint periodically issued to its sales force two documents: (1) a Master Incentive Compensation Guide (“MICG”), and (2) a Commissions Ac-knowledgement Form (“CAF”). The two documents together made up the employees’ compensation plan. During the class period, Sprint issued five different MICGs and 12 different CAFs, creating a total of 13 combinations of documents-in-force. These adhesion contracts were handed down from Sprint to its entire retail sales force — they were not negotiated. All of the contracts contain venue and choice of law provisions requiring that any lawsuits be filed in Kansas, and Kansas law applies.

Unfortunately, the advanced technology that allows Sprint’s cellular telephones to work may be easier to explain than the commission plans Sprint used to compensate its retail sales force. Sprint could have used a simple formula — say, a 5% commission on all sales above $1,000 per month, or $5 for each product and service sold — to motivate its employees to sell more. Instead, however, Sprint created labyrinthine commission plans requiring convoluted assessments. As explained later in this Report, Sprint relies heavily on this complexity in support of its decertification motion.

To illustrate this complexity, the Special Master provides a single example of a calculation required to determine only one portion of an employee’s commission compensation.

Generally, the MICGs set out overarching terms and definitions, while the CAFs set out more detailed provisions. For example, the 2006 MICG (which was actually in force for IB months, from May 1, 2006 to May 31, 2007) explains that: (i) an “Activation” occurs when “a new Sprint Customer activates service” and subscribes to a calling plan; (ii) a “Deactivation” occurs when “a new Sprint Customer voluntarily cancels service;” and (in) “Net Activations” refers to “Gross Activations in a given time period minus any Deaetivations within a given time period.” See docket no. 618-41 at 6-8 (setting out the full and more complicated definitions).⁵

The 2006 CAP then describes how an employee could receive commission, based on her Net Activations. Specifically, the 2006 CAF set out both a “quota” and a “target” for Net Activations, depending on the employee’s position — for example, the monthly quota for a “Retail Consultant” was 36, and the target was $350. See docket no. 618-29 at 116,109. If the employee sold more than her pionthly quota, she would receive a higher commission — for example, if Retail Consultant Andrea had 42 Net Activations, she would receive 42/35 x $350 = $420. Conversely, if Retail Consultant Beth did not meet quota, and had only 21 Net Activations, she would receive 21/35 x $350 = $210.

This initial calculation is relatively straightforward. In addition, however, each Retail Consultant would also receive an “accelerator” if she sold more than 100% of quota, and a “decelerator” if she sold less than 80% of quota. To continue the examples: the accelerator for Retail Consultant Andrea, who had 42 Net Activations, would be calculated as follows: ((42/35) x 1)-1) x $350 = $70. Thus, Andrea would get an accelerator of $70 in addition to her $420 for Net Activations, for a total of $490. In contrast, the decelerator for Beth, who had 21 Net Activations, would be calculated as follows: ((21/35) x 1)-1) x $350 =-$140. Thus, Beth would get a deeel-erator of $140 subtracted from her other $210 for Net Activations, for a total of $70.

Moreover, these “Net Activation” computations were only a small part of the total calculation necessary to determine Andrea’s and Beth’s monthly commissions. “Net Activations” was only one of the four commission components under the 2006 CAF; in addition, Retail Consultants also received commission for: (1) sales of services that created “Recurring Revenue,” such as a monthly data plan; (2) sales of products that created “Non-Recurring Revenue,” such as a phone charger; and (3) “Customer Satisfaction.” As with the “Net Activations” commission component, the 2006 MICG and CAF set out elaborate definitions for each of these other three components, and complex computations to assess the commission that flowed from each.⁶

The intricacy of Sprint’s compensation plans was compounded even further, for the following reasons:

• Sprint sometimes modified the definition of a given commission plan component over time. Three examples: (1) in 2006, sales reps suffered deductions from commissions flowing from Activations if there was a subsequent Deactivation within 14 days, while in 2007, Sprint lengthened this period to 180 days; (2) the list of products (“SKU list”) and list of services (“SOC list”) that were commission-eligible when calculating “Recurring Revenue” and “Non-Recurring Revenue” often changed throughout the class period; and (3) under the 2005 CAF, one component of sales reps’ commission was “Total Revenue,” but this component initially did not include revenue from sales of prepaid phones, while starting in September 2005 it did.

• Sprint changed the commission plan components themselves. For example, the components of the plan from August 2005 to April 2006 were: (1) Net Activations, (2) Total Revenue, (3) Contract Renewals, (4) Phone Upgrades and Swaps, and (5) Customer Satisfaction. In contrast, the components of the plan from July 2008 to September 2009 were: (1) Recurring Revenue, (2) Accessory Revenue, (3) Customer Satisfaction, and (4) Customer Retention. As reflected in the chart below, over the entire class period, Sprint’s compensation plans included 10 different commission components, only four or five of which applied at any given time.

Image

• Even within a single month of a single plan, different types of employees received different commissions. For example, the equations and factors necessary to calculate the commission for a Retail Store Manager were different from those necessary to calculate the commission for a Retail Consultant. Indeed, the Manager’s commission depended, on the amount of commissions earned and sales made by the employees underneath him (a concept referred to by Sprint as “roll-up”).

This overview provides a small taste of the difficulties involved in accurately calculating the monthly commissions Sprint owed to its over-30,000 class-member retail employees.

III. Sprint’s Data.

Stepping back for a moment, the fact that Sprint’s compensation plans are complicated does not by itself mean the parties’ experts cannot now calculate the exact, correct commissions for all of Splint’s retail employees. Ultimately, the task facing the experts — reconciling the amounts Sprint actually paid to each employee with the amounts it owed her — -is simply a great big math problem, susceptible to solution via computer programming. As was illustrated in 2002, when NASA first landed a probe on an asteroid, computers allow businesses to undertake dramatically complex calculations and achieve precise results — IF all of the necessary data and equations are available and correct.

Unfortunately, however, the uncontrovert-ed evidence shows that: (a) Sprint’s efforts at using computer programs and processes to contemporaneously calculate commissions were often maladroit; and (b) the Sprint data that remains available for the experts to now perform a post-hoc reconciliation is incomplete.⁷

Specifically, the evidence shows that Sprint’s problems began when it merged with Nextel in August of 2005. At the time, the point-of-sale computer system in Sprint’s stores was known as Retail Management System, or “RMS.” In addition, Sprint used a different software program to bill its customers, known as P2K. Nextel’s existing structure, however, was different — it used the Global Expense Reimbursement System (“GERS”) for its point-of-sale computer system, and Ensemble software for customer billing.

Upon merging, Sprint decided to use its own, existing RMS point-of-sale system in all stores — which meant the GERS computer terminals in the Nextel stores needed to be switched over to RMS. As for billing, however, Sprint decided to change from using its own, existing P2K software to Nextel’s Ensemble software — which meant all of Sprint’s existing customer billing needed to convert to Ensemble. Sprint called this project the “UBP conversion,” for “Unified Billing Platform.”

[[Image here]]

Internal contemporaneous business records reveal that the merging of Splint’s and Nextel’s systems, and the UBP conversion of the billing software, did not go smoothly. For example, in 2006, one year after the merger, Sprint’s Director of Sales Commissions wrote: “putting the legacy Nextel comp[ensation] plan structure into the legacy Sprint systems is akin to a ‘round peg in a square hole’ and vice versa. When the processes fail to provide the same level of quality outputs [e.g., accurate commission and compensation statements], we’re surprised/disappointed.” Docket no. 616-8 (SPRSIB-8962792). Sprint’s Controller had the same assessment, writing: “we have made these comp[ensation] plans way way way to [sic] complex .... [T]he current model ... is not working for hundreds of reasons but primarily because we can’t pay on things we can’t count.” Docket no. 616-12 (SPRSIB-8959937).

In 2007, two years after the SprintyNextel merger, Sprint still was not obtaining “quality outputs.” Sprint’s Director of Program Management, who was assigned to fix the ongoing commissions problem, wrote: “After six weeks of working on commissions, I found the first major problem. As a company we were migrating off of one billing system called P2K and onto another billing system called Ensemble. I found that every (that is a strong word but accurate) link between the order entry systems and Ensemble was defective. * * * Not a single [link] was working correctly to carry the essential elements necessary to correctly identify the agent who sold our products down to the billing platform and subsequently on to the commission systems.” Docket no. 616-7 (SPRSIB-11213902). At about the same time, Sprint’s Vice President of Business Performance Management, Dennis Peery, characterized things more succinctly: “commissions as an end-to-end process is broken and will probably take ‘Lawyers, Guns, and Money to fix in the long run.” Docket no. 616-31 (SPRSIB-9697043).

Sprint’s difficulty with accurately compensating its employees continued well after 2007. In February of 2008, Sprint’s Compensation Manager documented Splint’s ongoing problems as including several categories of missing data. See docket no. 616-36 (SPRSIB-7624457) (referring to eight data problem categories, one of which was being addressed by adding manually over 60,000 missing transactions every month to the database). In August of 2008 — a full three years after the merger — Sprint terminated Vice President Peery, even though he had not fixed the compensation system. Upon leaving, Peery wrote: “As for the Commissions work ... we simply have not demonstrated an ability to successfully implement commission systems or administer incentive sale comp[ensation] ourselves which accounts for our current situation. We just do need [sic, “not”] have all of the skills or the resources to fix and manage this and I believe that we absolutely need outside counsel, experience, and insight into leading industry practices to address these problems.” Docket no. 616-59 (SPRSIB-9722580).

Indeed, in March of 2009, close to four years after the merger, a Sprint Information Technology employee, who was assigned to fix the compensation system, observed: “[T]here are issues that are being worked today that are the same they were working over a year ago. The # 2 priority from one year ago [was] ‘Missing Add-on Revenue and Data Revenue’ and the number one priority for tody’s [sic, “today’s”] log is ’Missing add-ons for recurring revenue” .... This is a perfect example of working hard but making no progress. * * * [T]here has been zero improvement in the stability and overall availability of the system.” Docket no. 616-61 (SPRSIB-16468264).

In sum, the reason there remains a dispute in this case is not that it is impossible for the experts to reconcile the amounts Sprint paid to its retail employees with the amounts Sprint actually owed them. In fact, both sides’ experts did perform an employee-by-employee, month-by-month reconciliation for the entire 50-month class period.⁸ A dispute remains, however, mostly because the data available to calculate what Sprint actually owed is contradictory or incomplete, and the experts took different approaches to resolve discrepancies and “fill in the gaps.”⁹

IV. Class Certification Standards.

In its 2008 Certification Order, this Court set out the following standard for class certification, which remains potent today:

As the parties seeking class certification, plaintiffs have the burden to demonstrate “under a strict burden of proof’ that the requirements of Rule 23 are clearly satisfied. See Trevizo v. Adams, 455 F.3d 1155, 1162 (10th Cir.2006). In doing so, plaintiffs first must satisfy the prerequisites of Rule 23(a), that is, they must demonstrate that (1) the class is so numerous that joinder of all members is impracticable, (2) questions of law or fact are common to the class, (3) the claims of the representative parties are typical of the claims of the class[,] and (4) the representative parties will fairly and adequately protect the interests of the class. Rule 23(a), Fed. R. Civ. P. After meeting these requirements, plaintiffs must demonstrate that the proposed class action fits within one of the categories described in Rule 23(b). In this case, plaintiffs seek to proceed under 23(b)(3), which requires that “questions of law or fact common to the members of the class predominate over any questions affecting individual members” and that a class action “is superior to other available methods for the fair and efficient adjudication of the controversy.”

Sibley v. Sprint Nextel Corp., 254 F.R.D. 662, 670 (D.Kan.2008) (footnote omitted).

The Court further noted that, “[i]n deciding whether to certify, the Court must perform a ‘rigorous analysis’ whether the proposed class satisfies the requirements of Rule 23.” Id. at 669 (quoting General Tel. Co. of the Southwest v. Falcon, 457 U.S. 147, 155, 102 S.Ct. 2364, 72 L.Ed.2d 740 (1982)). The Supreme Court has recently reiterated this directive: “Repeatedly, we have emphasized that it may be necessary for the court to probe behind the pleadings before coming to rest on the certification question, and that certification is proper only if the trial court is satisfied, after a rigorous analysis, that the prerequisites of Rule 23(a) have been satisfied.” Comcast Corp. v. Behrend, — U.S. —, -—, 133 S.Ct. 1426, 1432, 185 L.Ed.2d 515 (2013) (internal quotation marks and citations omitted); see Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1233-34 (11th Cir.2016) (“A district court that has doubts about whether ‘the requirements of Rule 23 have been met should refuse certification until they have been met.’ ”) (quoting Fed. R. Civ. P. 23 advisory committee’s note to 2003 amendment). Nonetheless, “the Court should not conduct a mini-trial to determine if the class could actually prevail on the merits of their claims.” In re Motor Fuel Temperature Sales Practices Litig., 292 F.R.D. 652, 661 (D.Kan.2013) (citing Amgen Inc. v. Conn. Retirement Plans & Trust Funds, — U.S. —, —, 133 S.Ct. 1184, 1194-95, 185 L.Ed.2d 308 (2013), and Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 131 S.Ct. 2541, 2552 n. 6, 180 L.Ed.2d 374 (2011)).

When faced with a motion to decertify — that is, even after the Court has granted a motion for class certification — the party seeking certification continues to bear the burden of meeting the requirements of Rule 23. See Arkalon Grazing Ass’n v. Chesapeake Operating, Inc., 2014 WL 3089556 at *1 (D.Kan. July 7, 2014) (Murguia, J.) (“it remains the plaintiffs burden to prove that the requirements of Rule 23 are met”) (citing Wallace B. Roderick Revocable Living Trust v. XTO Energy, Inc., 725 F.3d 1213 (10th Cir.2013)).¹⁰

V. Sprint’s Arguments.

Of the six prerequisites listed in Rules 23(a) and (b)(3), Sprint does not contest four of them: numerosity, typicality, adequacy, and superiority. The Court’s earlier analysis of these four prongs is complete and convincing, so the Special Master does not touch on them in this Report. See Certification Order at 12-14 (numerosity), 16-18 (typicality and adequacy), 20-22 (superiority) (docket no. 99). See also Harlow v. Sprint Nextel Corp., 254 F.R.D. 418 (D.Kan.2008) (certifying a 23(b)(3) class of employees against Sprint).¹¹ Through its silence, Sprint has conceded that plaintiffs meet their burden of establishing these four criteria.

Sprint does insist, however, that plaintiffs do not carry their burden of establishing commonality or predominance. Sprint relies heavily on Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 131 S.Ct. 2541, 180 L.Ed.2d 374 (2011), which post-dates the Court’s original Certification Order, arguing that Dukes now mandates a different result.¹² Dukes focused on the commonality requirement, holding that the claims made by a plaintiff class must depend upon a “common contention ... [that] is capable of elasswide resolution— which means that determination of its truth or falsity will resolve an issue that is central to the validity of each one of the claims in one stroke.” Id. at 350 n. 6, 131 S.Ct. 2541. The Dukes court explained: “What matters to class certification ... is not the raising of common ‘questions’ — even in droves — but, rather the capacity of a elasswide proceeding to generate common answers apt to drive the resolution of the litigation.” Id. (quoting Richard A. Nagareda, Class Certification in the Age of Aggregate Proof, 84 N.Y.U.' L. Rev. 97, 132 (2009)). Sprint insists plaintiffs cannot meet the Dukes test of commonality, arguing:

The close of both fact and expert discovery reveal[s] that there is no common theory or common resolution to the question of whether Sprint, in fact, breached the terms of the commission contracts by underpaying each Named Plaintiff and class member. Indeed, each class member’s commissions are governed by contracts, which themselves are individualized, and by a commissions system and process involving a huge number of context-dependent factors (which take into account human discretion and human error). Unsurprisingly, at this late stage in the proceeding, Plaintiffs still can only offer a litany of vaguely phrased questions, none of which are common, masking what can only be a person-by-person, transaetion-by-transaetion determination of liability, which is untenable under Supreme Court and Tenth Circuit law. Thus, decertification is warranted.

Decertification Memo at 5 (docket no. 620).

Further, Sprint observes that “[t]he predominance criterion of Rule 23(b)(3) is ‘far more demanding than the Rule 23(a) commonality requirement,” In re Bank of Am. Wage & Hour Employment Litig., 286 F.R.D. 572, 577-78 (D.Kan.2012) (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)), and argues plaintiffs therefore cannot show predominance, either. In support, Sprint points to Comcast Corp. v. Behrend, — U.S. —, 133 S.Ct. 1426, 185 L.Ed.2d 515 (2013), another Supreme Court case that post-dates this Court’s original Certification Order.13 Comcast holds that class certification is inappropriate unless plaintiffs can “establish ] that damages are capable of measurement on a elasswide basis.” Id. at 1433. Without a methodology for doing so, plaintiffs “cannot show Rule 23(b)(3) predominance: Questions of individual damage calculations will inevitably overwhelm questions common to the class.” Id. Sprint insists the plaintiffs in this case “have not come forward with any legally valid or proeedurally viable mechanism for proving that every class member in fact suffered a net loss in commission payments[, as required to prove] their breach of commission contract claim against Sprint. Rather, as illustrated by Plaintiffs’ own experts’ analysis, the determination of underpayments is an individualized process, which will consequently yield a massive number of individualized and contentious mini-trials.” Decertification Memo at 23 (docket no. 620) (citation omitted).

Ultimately, Sprint’s logic runs like this: (1) to prove her claim for breach of contract, as well as her claim for violation of the KWPA, each plaintiff must prove she suffered damages;¹⁴ (2) to prove she suffered damages, each plaintiff must (a) identify all of her own, individual commissionable transactions (of which there will be many thousands), (b) calculate the commissions she earned (which will require massive and intricate computations), and then (c) compare her commissions earned with her commissions received, to determine if she was underpaid (requiring yet additional calculations); and (3) this analysis must be done for each individual sales rep — it cannot be done class-wide. Indeed, Sprint notes, even plaintiffs’ own experts conclude that about 4,000 of the over 30,000 class members were not underpaid¹⁵ —which means they were not damaged, which means they can’t prove their claims, which means they were never properly a part of the class in the first place, which shows the class should never have been certified. In sum, Sprint insists the class is not “sufficiently cohesive to warrant adjudication by representation.” Amchem, 521 U.S. at 623, 117 S.Ct. 2231 (describing the “predominance inquiry”).

For the reasons stated below, the Special Master concludes Sprint’s arguments fail.

YI. Analysis.

A. Commonality.

Sprint’s attack on commonality is two-pronged. First, Sprint argues plaintiffs cannot “establish commonality under the terms of the commission contracts;” and second, Sprint argues plaintiffs “cannot establish commonality within Sprint’s commissions processes, systems, and data.” Decertificar tionMemo at 11,15 (docket no. 620). Neither of these prongs withstands analysis.

1. Commonality of Contracts

In an effort to prove plaintiffs cannot cany their burden of showing “there are questions of law or fact common to the class,” Fed. R. Civ. P. 23(a)(2), Sprint focuses on the complexity of its commission contracts with its sales reps. As described earlier, Sprint changed its commission contracts numerous times and in numerous ways over the four-year class period. These changes were both broad (for example, Sprint substituted entire “commission components,” such as its 2006 discontinuation of commission based on “Contract Renewals” and initiation of commission based on “Recurring Revenue”), and also specific (for example, in 2006, Sprint changed the quota for a Retail Consultant’s Net Activations from 50 to 35).

Accordingly, Sprint observes that plaintiffs are “ehalleng[ing] separate commissions calculations for millions — perhaps billions — of individual transactions subject to different commission contracts and the varying terms therein.” Decertification Memo at 14 (docket no. 620). Sprint asserts “these contractual variations for each individual and for his/her transactions render any resolution as to the question of whether Sprint breached the commission contract incapable of classwide resolution. In other words, at this stage of the litigation, the commission contracts themselves cannot serve as the ‘glue’ to hold Plaintiffs’ claims together — nor is there any other ‘glue’ available in this case.” Id.

In making this argument, Sprint relies on the following portion of the Supreme Court’s reasoning in Dukes:

In this case, proof of commonality necessarily overlaps with [plaintiffs’] merits contention that Wal-Mart engages in a pattern or practice of discrimination. That is so because, in resolving an individual’s Title VII claim, the crux of the inquiry is “the reason for a particular employment decision,” Cooper v. Federal Reserve Bank of Richmond, 467 U.S. 867, 876, 104 S.Ct. 2794, 81 L.Ed.2d 718 (1984). Here [plaintiffs] wish to sue about literally millions of employment decisions at once. Without some glue holding the alleged reasons for all those decisions together, it will be impossible to say that examination of all the class members’ claims for relief will produce a common answer to the crucial question why was I disfavored.

Dukes, 564 U.S. at 352, 131 S.Ct. 2541 (footnote omitted, emphasis in original).

What Sprint overlooks, however, is that the “crux” of the Dukes plaintiffs’ claims was intent — that is, all of the individual “reasons” for all of Wal-Mart’s “employment decisions” — which has nothing to do with plaintiffs’ claims in this case. Sprint’s intent is irrelevant to whether it breached its contracts with plaintiffs.¹⁶ Moreover, the intent of Wal-Mart’s managers when they made decisions on whether to promote was highly personalized, while the “decisions” of Sprint’s computer system on whether a sale rep earned a commission were literally programmed.

Rather, the crux of the plaintiffs’ claims in this case are the terms of their commission contracts and Sprint’s performance vel non of its obligations thereunder, using its computerized commission-calculation system. Even though there are over 30,000 class members, there was only one contract that applied to all employee class members at any given time. Each contract was complex, but each contract was common to every plaintiff in this case during her employment. Further, Sprint used a single automated process to calculate amounts it owed to all of the class members under all of the contracts.¹⁷ Again, this process was complex, but it applied equally to every class-member plaintiff. While each plaintiff engaged in different eommissionable transactions, the crux of each plaintiffs claims — the applicable contract terms regarding commissions owed, and Sprint’s system for identifying and paying those commissions — is entirely identical over time.

Indeed, Sprint attempts to borrow too heavily from the reasoning in Dukes when it asserts that “Plaintiffs must affirmatively demonstrate with sufficient evidentiary proof that that [sic] there is common proof to answer the question of liability: whether Plaintiffs suffered a net underpayment of wages (i.e., were they damaged), and why that underpayment occurred.” Decertification Memo at 12 (docket no. 620) (emphasis added); see also id. at 17 (“Plaintiffs still have no common proof that answers the critical question as to why they were underpaid”) (emphasis added). To repeat, the elements of plaintiffs’ contract claims do not require proof of intent, nor the “reason why” they were underpaid. See City of Andover v. Southwestern Bell Tel., L.P., 37 Kan.App.2d 358, 153 P.3d 561, 565 (2007) (setting out the elements of a claim for breach of contract). Numerous cases have noted that analogies to Dukes fail if the plaintiffs’ claims “do not require an examination of the subjective intent behind millions of individual employment decisions.” Youngblood v. Family Dollar Stores, Inc., 2011 WL 4597555 at *3 (S.D.N.Y. Oct. 4, 2011). Rather, certification is appropriate if — as in this case — the principal issue “is whether the company-wide policies, as implemented, violated Plaintiffs’ [contractual] rights.” Id. (certifying FLSA claims); see Buchanan v. Hamos, 2012 WL 1953146 at *9 (N.D.Ill. May 30, 2012) (“as the Seventh Circuit has recognized, the holding in Dukes does not apply where discrimination results from a defendant’s standardized conduct toward proposed class members, such as generalized policies that affect all class members in the same way.”) (emphasis added) (citing McReynolds v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 672 F.3d 482, 487 (7th Cir.2012)).

Sprint’s computerized system for compensating plaintiffs represents “standardized conduct toward proposed class members” that affected them all in the same way. Id. Sprint disputes this point, noting that each plaintiff will have to offer different reasons for her alleged underpayment, depending on her particular circumstances:

To be sure, breach of contract claims do not require proof of intent, nor proof of why or how the breach came about; but if Plaintiffs cannot explain how the commissions systems allegedly were “broken,” or what underlying problems regularly and mechanically brought about alleged underpayments with respect to what terms under which commission contracts, * * * then there are no common questions, no “glue” holding the Plaintiffs’ claims together, and no way to prove liability with common evidence. If, for example, the commissions systems systematically deducted a dollar from every commission for a particular component for a particular commission contract year, proof of that problem would have a common effect on all class members. Plaintiffs cannot point to any uniform underlying issue, however; instead they * * * [refer to a] shifting “myriad of reasons” that contribute to any particular commission calculation for a particular individual under a particular commission contract — [which] prevents any evaluation of Sprint’s liability to all class members based on common evidence that all would share.

Sprint Reply at 16 (docket no. 632). That this argument fails, however, is illustrated by Sprint’s own hypothetical. Imagine that, instead of “systematically deducting] a dollar from every commission for a particular component for a particular commission contract year,” id., an unidentifiable bug in Sprint’s commission software code worked to subtract random amounts from random commissiona-ble events from random sales reps at random times. Sales reps who tracked their own performance would be able to show they were underpaid by looking at historical data, but would be unable to identify why. Even though the plaintiff class would share no single, common commission-calculation problem, the problematic commissions system “would have a common effect on all class members” — their commissions would be incorrectly low. Id.

Moreover, to adopt Sprint’s position would be to reward ineptitude — if Sprint’s commission calculations are wrong because of a single “underlying issue” (such as “systematically deduct[ing] a dollar” from every commission owed for the sale of a texting plan), then Sprint concedes certification is appropriate; but if Sprint’s calculations are wrong because of numerous computer problems (such as lost data, faulty calculations, missing data-links, and incorrect programming), then certification is not appropriate. The commonality analysis is not that granular. The “single underlying, common issue” in this case is that Sprint’s commission-calculation system did not work.

Finally, it is worth noting that Sprint is not correct when it asserts plaintiffs “cannot explain how the commissions systems allegedly were ‘broken,’ or what underlying problems regularly and mechanically brought about alleged underpayments with respect to what terms under which commission contracts.” Id. Plaintiffs’ experts have identified dozens of ways that Sprint’s commissions system was allegedly broken. That these many problems are interdependent and interactive make it difficult to trace exactly why a given monthly commission check to a given sales rep was incorrect, just as would be the case for the random-number generator hypothetical. But plaintiffs’ experts have stated explicitly why they believe Sprint’s commission computations were incorrect, and have offered examples both minute and overarching.¹⁸

This is not to say that classes bringing contract claims should always be certified simply because the reason for breach is not an issue. For example, Sprint cites three related cases where certification of a class that asserted claims for breach of contract was vacated. But the reason in each case was that the defendant had entered into meaningfully different contracts with different members of the class. See Wallace B. Roderick Revocable Living Trust v. XTO Energy, Inc., 725 F.3d 1213, 1216-19 (10th Cir.2013) (plaintiffs’ lease contracts contained “language variations” yielding “twenty different categories of royalty provisions,” and the district court had not even “create[d] a chart classifying lease types”); Chieftain Royalty Co. v. XTO Energy, Inc., 528 Fed.Appx. 938, 941-42 (10th Cir.2013); (noting class members’ leases contained “86 different royalty clauses” and “the district court did not examine whether lease language variations destroy the possibility of resolving the common question on a classwide basis”); Arkalon Grazing Ass'n v. Chesapeake Operating, Inc., 2014 WL 3089556 at *3 (D.Kan. July 7, 2014) (Murguia, J.) (decertifying a class because “the court — after considering again the lease language variations and the question of marketability as directed by the Tenth Circuit in Roderick and Chieftain — finds that individual inquiries into the language of each lease and the marketability of gas at each well preclude a finding of commonality”) (citations omitted). In contrast, and as noted earlier, in this case there was only one operative contract at any given time during the class period, and it applied to every single employee class member. The different circumstances in Roderick, Chieftan, and Arkalon make them inappo-site.¹⁹

In sum, that the Sprint commission contracts are complex does not make their application to the sales reps uncommon. The same contractual provisions applied to every employee class member at any given time, and Sprint used the same computerized system to calculate commissions automatically for every class member. It is clear “there are questions of law [and] fact common to the class,” Fed. R. Civ. P. 23(a)(2), and there are numerous common contentions that can be resolved (via motion or verdict) in a single stroke.

2. Commonality Within Sprint’s Commissions Processes, Systems, and Data

In addition to arguing that plaintiffs cannot show there are common questions related to the commission contracts, Sprint also argues plaintiffs cannot show there are common questions related to Sprint’s computerized commission-calculation system. This argument also fails, for essentially the same reason — that is, Sprint views the facts through a lens that has an excessively narrow focus.

To begin, Sprint repeats an argument the Court has already rejected more than once. When this Court certified the class six years ago, it noted that the “class members all claim that defendants systematically denied them commissions through under-reporting of sales and improper deductions from commissions.” Certification Order at 14 (docket no. 99). Sprint today insists the plaintiffs “have abandoned their original theories for certification and [now] claim that due to ‘numerous’ unspecified problems, Sprint failed to pay ‘accurate commissions to the class, routinely underpaying them for the commissions they earned.’ ” Decertification Memo at 15 (docket no. 620). The change, according to Sprint, is that plaintiffs originally asserted there was a single problem that systematically caused underpayment of commissions to all class members, while plaintiffs now assert there are numerous problems that routinely cause inaccurate payment of commissions to most class members. Sprint also insists plaintiffs “have failed to show that any systematic problem affected all class members — let alone identify such problem.” Id. at 16.

But Sprint is wrong — plaintiffs never asserted Sprint’s computerized commission-calculation system suffered only a single, discrete problem, and three Judges of this Court have expressly recognized as much. Even in their original complaint, plaintiffs alleged that, “due to problems with Defendants’ computer systems, the Named Plaintiffs and the proposed class members were customarily and regularly denied the commissions Defendants promised to pay them.” Complaint at 5, ¶ 17 (docket no. 1) (emphasis added). And when this Court granted class certification, it understood there were at least two alleged system-wide problems— “under-reporting of sales and improper deductions from commissions” — which were in turn caused by numerous “issues ranging] from[: (i) ] substantial problems which affect a large number of employees to [ (ii) ] less significant, discrete issues that affect a relatively small number of employees.” Certification Order at 7-8 (docket no. 99) (emphasis added).

Moreover, in the companion Harlow case (see footnote 11 above), Judge Lungstrum observed that “Plaintiffs allege a systemic failure of underreported sales and widespread computer problems that resulted in unpaid commissions.” Harlow v. Sprint Nextel Corp., 254 F.R.D. 418, 424 (D.Kan.2008) (emphasis added). And yet a third judicial officer has ruled that plaintiffs’ theory was never limited to a single, discrete problem: when Magistrate Judge O’Hara granted plaintiffs’ motion to compel in this ease, he wrote: “[Sprint’s] argument appears to be that plaintiffs are not permitted to seek discovery into defendants’ payment or underpayment of commissions that are not caused by a single systematic computer failure. [Sprint’s] argument is rejected. Plaintiffs’ theory in this case is — and always has been — that problems (plural) in defendants’ computer systems led to the underpayment of commissions to class members.” Order at 3 (docket no. 279) (original paragraphing and footnote omitted; emphasis added). Put simply, plaintiffs’ theory of the case has been constant from the beginning, and it is clear the Court understood that theory when it certified the class.²⁰

In addition to rejecting Sprint’s assertion that plaintiffs have changed their thesis, the Special Master also rejects Sprint’s contention that the presence of numerous alleged data and system problems undermines commonality. As noted earlier, all of Sprint’s commission contracts, complex though they may be, applied equally to each class member at any given moment. Similarly, Sprint’s computerized commission-calculation system, convoluted though it may be, applied equally to each class member. The system was designed to mathematize the terms of the underlying contracts, and to apply the same formulae to all sales reps without distinction. Thus, a “math error” also affected all sales reps without distinction. For example, the alleged problem that use of multiple employee-identifiers caused the system to miss data-matches, and thus to fail to pay commissions, would affect every employee who had multiple identifiers — which is every class member. Similarly, if the system assigned an incorrectly low commission half the time for the sale of an “Everything Family Data Plan,” then every sales rep who sold that service would (on average) get the wrong commission half the time. Sprint is correct that the exact constellation of inaccurate commissions a given sales rep actually experienced is individualized — one rep might get the wrong low commission for the Everything Family Data Plan 195 times, while another might get the wrong commission 418 times, and yet another might get penalized zero times — but the essential nature of the problem, and its cause and effect, remain common. Ultimately, whether and the extent to which each alleged problem existed in Sprint’s computer system, and whether and the extent to which it affected the class, are central, common questions of fact for a jury to determine.²¹

Finally, Sprint argues that, “[e]ven if Plaintiffs were able to present common proof of systematic computer problems ..., Sprint’s individualized defenses to Plaintiffs’ breach of contract claim ... still defeat a finding of commonality.” Decertification Memo at 20 (docket no. 620) (citing Tabor v. Hilti, Inc., 703 F.3d 1206, 1229 (10th Cir.2013)). This argument fails for several reasons. First, Sprint cites only one affirmative “individualized” defense: an “express contractual provision” stating that “each class member had a responsibility to ... appeal any underpayments as a condition of payment.” Id. Given that this contractual provision applied equally to every class member, however, and that plaintiffs have moved for summary judgment on this precise issue across the entire class,²² this “individualized defense” weighs about as much in favor of a finding of commonality as against it. Second, in Tabor, the Tenth Circuit Court of Appeals affirmed denial of class certification because the plaintiffs did “not show[ ] that [the defendant employer] maintained ‘a common mode of exercising discretion that pervade[d] the entire company.’” Tabor, 703 F.3d at 1229 (quoting Dukes, 564 U.S. at 356, 131 S.Ct. 2541). Here, Sprint employed a ubiquitous computerized system that only rarely called for exercise of discretion; Tabor is instructive because of its contrast.²³

And third, this Court earlier observed that “[a] statute of limitations defense does not prevent certification ... if the appropriate Rule 23 factors ... are otherwise met.” Certification Order at 14 n.13 (docket no. 99) (citing cases); see also Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1240 (11th Cir.2016) (“individual affirmative defenses generally do not defeat predominance”); Smilow v. Southwestern Bell Mobile Sys., Inc., 323 F.3d 32, 39 (1st Cir.2003) (in a case brought by a class of cellular telephone customers asserting breach of contract claims, observing that “Courts have traditionally been reluctant to deny class action status under Rule 23(b)(3) simply because affirmative defenses may be available against individual members.”); Nieberding v. Barrette Outdoor Living, Inc., 302 F.R.D. 600, 617 (D.Kan.2014) (Crabtree, J.) (concluding that, even though the statute of limitations defense “requires some individualized inquiries,” that “does not mean that individual issues predominate over common issues”). Similarly, the contract defense Sprint interposes does not destroy commonality nor mandate decertification.

In sum, that Sprint’s computerized commission-calculation system is complicated, and that the data the system processed is copious, does not make the system’s relationship to each sales rep uncommon. The system was designed to apply the same complex formulae to every class member automatically. Whether it failed to pay commissions owed is a question common to each member of the class, and a jury can determine a global answer.

B. Predominance.

Despite the Special Master’s conclusion that there are numerous questions of law and fact common to the class, Sprint is certainly correct that there are also factual questions that are individual to each plaintiff class member. The principal individualized fact question is damages: how much of a commission underpayment, if any, did each class member suffer? Answering this question requires reconciling the amounts Sprint paid to each employee with the amounts it owed her. It is easy to ascertain the former amount (commission payments actually made), but calculating accurately the latter amount (commissions owed) is exceedingly difficult. The parties’ experts agree the only way to undertake this reconciliation is by: (1) constructing a massively mulitplex computerized model that translates the terms of the commission contracts into mathematical for-mulae; and (2) running all of the billions of relevant data points through the model to yield month-by-month, employee-by-employee comparisons. The relevant commissionable transactions, and the commission amounts actually paid, are idiosyncratic for each employee; and so is the final step in the computation, which is calculation of damages (if any) on an individualized basis.

Sprint maintains the idiosyncracies are overwhelming: “the actual trial process will involve at least 30,000 excruciatingly lengthy and individualized mini-trials, each of which will involve a battle between the experts to reconcile each individual transaction of each class member.” Decertification Memo at 24 (docket no. 620). Sprint warns the jury will have to:

review the terms of the applicable commission contract (which changed year-to-year), determine whether the transaction(s) in question constituted a Commissionable Event, whether the individual was paid for the transaction(s), and, if not, whether there was a match of the sales and billing records in the commissions system and whether the employee chose to appeal the non-payment as required under the compensation plans. As shown at the conclusion of fact and expert discovery, these questions cannot be answered with common proof, but instead require an individualized review of each of millions of transactions and the eligibility for each class member to receive commissions.

Id. Accordingly, Sprint insists plaintiffs fail to carry their burden of showing that “questions of law or fact common to the members of the class predominate over any questions affecting individual members,” as required by Fed. R. Civ. P. 23(b)(3). See Roderick, 725 F.3d at 1220 (“We note that Rule 23(b)(3)’s predominance criterion is ‘far more demanding* than Rule 23(a)’s commonality requirement.”) (quoting Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 623-24, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997)).

Sprint’s concern is legitimate. In Roderick, the Tenth Circuit Court of Appeals vacated class certification and directed the lower court, on remand, to reassess plaintiffs’ motion to certify. The Circuit court suggested the Kansas district court (Marten, J.) should give greater consideration to “the extent to which material differences in damages determinations will require individualized inquiries.” Id. Noting that “predominance may be destroyed if individualized issues will overwhelm those questions common to the class,” the appellate court also counseled the district court to “consider ‘how a trial on the merits would be conducted if a class were certified.’” Id. (quoting Gene & Gene LLC v. BioPay LLC, 541 F.3d 318, 326-29 (5th Cir.2008)).²⁴

The Special Master has given considerable thought to the question of predominance, and particularly the question of how a trial on the merits would go forward. Ultimately, the Special Master is persuaded by plaintiffs’ observation that the jury can assess the experts’ computerized reconciliations in order to determine damages and need not (indeed, cannot) undertake a manual transaction-by-transaction, employee-by-employee analysis. As plaintiffs explain:

Both parties’ experts have already reconciled each Plaintiffs damages in a “transaction-by-transaction” analysis that was performed in an automated, class wide manner. Both parties’ experts agree this is the best and most accurate way to determine if Sprint underpaid its sales employees. Both parties agree that any error in either expert’s calculation applies to the entire class. And Sprint similarly paid the class in the same fully automated, person-by-person, transaction-by-transaction manner that Splint now claims is not good enough from Plaintiffs’ experts.

In other words, Sprint claims Plaintiffs have to do more to prove a breach of the contract than Sprint was supposed to do to comply with it.

Plaintiffs’ Response at 47 (docket no. 626) (original emphasis deleted).

Indeed, that plaintiffs are correct is proved by imagining the trial for a single class-member. Even then, neither Sprint nor the plaintiff could possibly undertake a manual, transaction-by-transaetion analysis of the lone plaintiffs commission earnings. Rather, both sides would have to create a computerized commission-calculation model, just as Sprint did originally in order to pay its employees, which mathematizes the applicable contracts and processes the relevant data. Moreover, the parties’ lone-plaintiff computerized models would be about equally complex as their current class-wide models — the lone-plaintiff models would still have to incorporate virtually all of the same formulae, achieve all of the same software links, process all of the same types of information, accommodate all of the same missing-data issues, and so on. If the lone-plaintiff is a manager, the computerized model would even have to process data from other class-member employees, since a manager’s commission depends on his employees’ commis-sionable performance. Indeed, because the evidence shows some transactions were not linked to any employee, and the experts could only match these orphan transactions to the correct sales rep through additional investigation, the only way for even the lone plaintiff to capture every commission owed is to examine the entire class-wide data set. And Sprint’s expert would doubtless still assert her reconciliation is “more accurate than Sprint’s” original, contemporaneous commission calculations. Depo. at 139 (docket no. 626-18).

Ultimately, the essential assessments a jury (and the Court) would have to undertake at trial for a lone class-member plaintiff would largely mirror those a jury (and the Court) will have to make for the entire class. The jury’s assessments in either ease will be the extent to which each expert’s commission-calculation model accurately mathema-tized the applicable contracts, processed the relevant data, and produced accurate results.²⁵, ²⁶

Notably, case law supports certification in various types of cases where there are individualized damages issues. For example, “[t]he individuation of damages in consumer class actions is rarely determinative under Rule 23(b)(3).” Smilow, 323 F.3d at 40 (reversing decertification of a class of cellular telephone customers asserting breach of contract claims, even though they had different calling plans and individual damage amounts). Similarly, in a securities class action, “[t]he amount of damages is invariably an individual question and does not defeat class action treatment.” Blackie v. Barrack, 524 F.2d 891, 905 (9th Cir.1975) (affirming class certification). And in an antitrust case, “[ejommon issues may predominate when liability can be determined on a class-wide basis, even when there are some individualized damage issues.” In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 139 (2nd Cir.2001) (affirming class certification, and adding that “failure to certify an action under Rule 23(b)(3) on the sole ground that it would be unmanageable is disfavored”) (Sotomayor, J.).²⁷

See also 2 Newberg on Class Actions § 4:54 at 210 (5th ed. 2012) (“[OJutside of the personal injury/mass tort context, cases in which individual damage concerns predominate and defeat certification ‘rarely, if ever, come along.’”) (quoting Klay v. Humana, Inc., 382 F.3d 1241, 1260 (11th Cir.2004)); 1 McLaughlin on Class Actions ¶ 4:19 (12th ed. 2015) (“The overwhelming majority of cases addressing the effect of variations in the amount of damages recoverable by class members on the certification analysis have concluded that if there is a sufficient nucleus of common questions concerning liability, the fact that damages will need to be calculated on an individual basis does not defeat typicality or otherwise bar class certification”) (citing cases); 5 Moore’s Federal Practice § 23.23[2] (1997) (“[T]he necessity of making an individualized determination of damages for each class member generally does not defeat commonality.”); Comcast Corp. v. Behrend, — U.S. —, —, 133 S.Ct. 1426, 1437, 185 L.Ed.2d 515 (2013) (“Recognition that individual damages calculations do not preclude class certification under Rule 23(b)(3) is well nigh universal.”) (Ginsburg and Breyer, JJ., dissenting).

Further, “[c]ommon issues predominate where individual factual determinations can be accomplished using computer records, clerical assistance, and objective criteria — thus rendering unnecessary an evi-dentiary hearing on each claim.” Smilow, 323 F.3d at 40. Newberg explains as follows:

[T]he proponents of class certification [must] provide “a common methodology to measure and [2] quantify damages on a class-wide basis” or, as the Supreme Court recently put it, to establish “that damages are capable of measurement on a classwide basis.” [Comcast, 133 S.Ct. at 1433.] These phrases are unfortunate (though oft-repeated) in that the use of the word “class-wide” implies that there must be a measure of damages suffered by the whole class, so-called aggregate damages. In fact, the predominance inquiry is focused on something else entirely, namely that there must be a single or common method that can be used to measure and quantify the damages of each class member, lest individual damage calculations predominate over common questions of liability. The class proponents’ task, therefore, is to demonstrate a method for quantifying individual damages that applies across the board and hence is common to the class: a common classwide method for calculating individual damages.

4 Newberg on Class Actions § 12:4 (5th ed. June 2016 update) (emphasis in original, footnotes omitted). This “common classwide method for calculating individual damages” will often require “fancy models.” Id. That is exactly what plaintiffs have done here — their experts supply a “fancy,” computerized, commission-calculation model to quantify individual damages for each class-member plaintiff. (Splint has done essentially the same thing, although the two models reach very different results.) That plaintiffs have done so works to satisfy the predominance requirement. See Smilow, 323 F.3d at 40 (suggesting predominance is met if “plaintiffs could use a computer program to extract from Cellular One’s computer records information about individual damages”); id. at 42 (“Since [plaintiff] can compute actual damages using a computer program, we need not address [the] argument that common issues would predominate only on a claim for statutory damages, and not on a claim for individual damages.”).²⁸

Other courts have also specifically relied upon the availability of a computerized model of damages to conclude that predominance is met, despite individualized damage issues. See Leyva v. Medline Indus. Inc., 716 F.3d 510, 514 (9th Cir.2013) (overruling the denial of class certification and rejecting the trial court’s “manageability concerns” — specifically the worry that “the damages inquiry will be highly individualized” — because the defendant’s “computerized payroll and timekeeping database would enable the court to accurately calculate damages and related penalties for each claim”); Parra v. Bashas’, Inc., 291 F.R.D. 360, 393 (D.Ariz.2013) (certifying a class of grocery store employees because their damages model used “a computer program, and ... objective factors such as the individual employee payroll record ... to calculate back pay losses for each eligible class member”); In re Checking Account Overdraft Litig., 281 F.R.D. 667, 673 (S.D.Fla.2012) (certifying a class of bank customers who alleged they were improperly charged overdraft fees, noting the plaintiffs’ expert “is able to calculate damages on an account-by-account basis using the bank’s own computerized records”); Smith v. MCI Telecommunic’ns Corp., 124 F.R.D. 665, 677 (D.Kan.1989) (O’Connor, C.J.) (certifying a class of commissioned sales reps of a wireless carrier, because commissions owed to each rep could be calculated by “run[ning] the unpaid sales against billing data using ICS [the defendant’s computerized ‘Integrated Commission System’] with a special test program”).²⁹

In sum, the predominance requirement is satisfied “if resolution of some of the legal or factual questions that qualify each member’s case as a genuine controversy can be achieved through generalized proof, and if these particular issues are more substantial than the issue subject only to individualized proof.” In re Bank of Am. Wage & Hour Employment Litig., 286 F.R.D. 572, 577-78 (D.Kan,2012) (quoting In re American Int’l Group, Inc. Sec. Litig., 689 F.3d 229, 240 (2nd Cir.2012)). Plaintiffs meet that test here; there are common questions of fact and law that can be decided “in one stroke,” and these common questions are not overwhelmed by the individualized fact question of damages.

C. Other Issues.

Before closing, the Special Master confronts a few other issues not yet addressed.

1. Class Members Who Suffered No Damages.

Sprint notes correctly that a cause of action for breach of contract, under the applicable law of Kansas, includes damages as an element of the cause of action.³⁰ Thus, Sprint asserts — and plaintiffs agree — that to prove liability, you must prove damages. See Plaintiffs’ Response at 29 (docket no. 626) (“elements four and five [breach and damages] are mutually dependent, proof of one is proof of the other”). Sprint thus concludes that both liability and damages must be assessed on an individualized basis, so class treatment is inappropriate.

This argument fails for three reasons. First, the same assertion applies to each of the faetually-similar cases cited above, yet these courts all granted class certification. See, e.g., Harlow v. Sprint Nextel Corp., 254 F.R.D. 418 (D.Kan.2008); Smilow v. Southwestern Bell Mobile Sys., Inc., 323 F.3d 32 (1st Cir.2003); Heartland Communic’ns, Inc. v. Sprint Corp., 161 F.R.D. 111 (D.Kan.1995); Smith v. MCI Telecommunic’ns Corp., 124 F.R.D. 665 (D.Kan.1989).

Second, given that so many eases and treatises agree that the need for individualized damage determinations does not defeat commonality or predominance, it follows that the fact that some class members suffered no damages also does not defeat commonality or predominance. See, e.g., In re Urethane Antitrust Litig., 2013 WL 2097346 at *2 (D.Kan. May 16, 2013) (Lungstrum, J.), affirmed, 768 F.3d 1246 (10th Cir.2014) (“[Defendant] has not cited any authority supporting the argument that the presence of a few ‘zero-damages’ class members necessarily defeats certification. In fact, caselaw is to the contrary. For instance, the Seventh Circuit has noted that a class will almost inevitably include persons who have not been injured by the defendant’s conduct, and that fact (or even inevitability) does not preclude certification.”) (citing Northshore Univ. HealthSystem, 669 F.3d 802, 823 (7th Cir.2012); Gunnells v. Healthplan Servs., Inc., 348 F.3d 417, 428 n. 5 (4th Cir.2003) (affirming certification in part, even though defendant could “show that an individual class member did not suffer any damages”); Bouaphakeo v. Tyson Foods, Inc., 765 F.3d 791, 805 (8th Cir.2014) (Beam, J,, dissenting) (disagreeing with the majority’s decision to affirm certification, even though “significant numbers of the putative classes suffered] no injury and members of the entire classes suffered] wide variations in damages, ultimately resulting in a single-sum class-wide verdict from which each purported class member, damaged or not, will receive a pro-rata portion of the jury’s one-figure verdict”) (emphasis added), affirmed, — U.S. —, 136 S.Ct. 1036, 194 L.Ed.2d 124 (2016).³¹

And third, there is a supremely practical solution to the “problem” Sprint identifies in this case — if plaintiffs prevail, the class members whom plaintiffs have identified as “not-net-underpaid” will receive no compensation. By identifying before trial class members who do not prove liability cum damages, plaintiffs essentially tailor the class to meet Sprint’s objection.³²

2. Trial Complexity and the Jury.

A theme accurately presented by Sprint is that the experts’ models, and the individualized calculations they undertake, are extraordinarily complex. Having examined these models in great detail, the Special Master and the Technical Advisor both join the Court in its concern that “a jury will find the experts’ opinions so opaque that the jury will be forced simply to guess” who is correct. Appointment Order at 2 (docket no. 532). If the parties agree to a bench trial, this concern would be mitigated. If not, it will be the parties’ job to make their experts’ models, and their attacks on the opposing experts’ models, understandable to the jury.

What is certain, however, is that “[cjertification cannot be denied [simply] because the number of potential class members makes the proceeding complex or difficult.” In re Theragenics Corp. Sec. Litig., 205 F.R.D. 687, 697 (N.D.Ga.2002). “Difficulties in management of a case become significant only if they make the class action a less fair and efficient method of adjudication than other available techniques.” Id. at 697-98 (citations and internal quotation marks omitted).

This Court wrote earlier it was “persuaded that a class action is the superior method for resolving the claims at issue in this lawsuit,” and Sprint has not challenged this conclusion.³³ Appointment Order at 2 (docket no. 532). The Special Master now concurs, especially because the experts’ commission-calculation models for the trial of a lone plaintiffs claims would be almost equally complex as the models presented in this class action. See CGC Holding Co., LLC v. Broad & Cassel, 773 F.3d 1076, 1096 (10th Cir.2014) (“It is enough that class treatment is superior because it will ‘achieve economies of time, effort, and expense, and promote uniformity of decision as to persons similarly situated, without sacrificing procedural fairness or bringing about other undesirable results.’ ”) (quoting Amchem, 521 U.S. at 615, 117 S.Ct. 2231).

If there is comfort to be had, it lies in the long history of juries tackling and resolving cases where opposing experts opine on extremely esoteric matters. See, e.g., In re Urethane Antitrust Litig., 2013 WL 2097346 (D.Kan. May 15, 2013) (Lungstrum, J.), affirmed, 768 F.3d 1245 (10th Cir.2014) (jury accepted in part and rejected in part plaintiffs’ expert’s abstruse economic model of antitrust damages); In re Welding Fumes Prods. Liab. Litig., 2010 WL 7699456 (N.D.Ohio June 4, 2010) (in different bellwether trials, juries relied on expert testimony to reach different conclusions regarding the type and cause of plaintiffs brain damage);³⁴ Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 596, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993) (rejecting views that were “overly pessimistic about the capabilities of the jury and of the adversary system generally”).

In the end, the computerized commission-calculation models offered by the experts in this case are of the same ilk as multiple-regression analysis models in antitrust cases, or computer programming code comparisons in patent cases. Thus, complexity, alone, does not weigh in favor of decertification.

3. Damages Determinations.

To some extent, class action case law addressing the issue of individual damages just “kicks the can down the road.” Courts commonly explain that class certification can be appropriate even if individual damage determinations will be required, and then observe there are several “[case] management tools” a court can use once actual trial rolls around. In re Visa Check/MasterMoney Antitrust Litig., 280 F.3d 124, 141 (2nd Cir.2001). These tools include, but are not limited to, “(1) bifurcating liability and damage trials with the same or different juries; (2) appointing a magistrate judge or special master to preside over individual damages proceedings; (3) de-certifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; (4) creating subclasses; or (5) altering or amending the class.” Id.

At this juncture, however, there is not much road left for can-kicking. Thus, the Special Master addresses how this trial of this case might actually go forward. See Roderick, 725 F.3d at 1220 (the court must “consider ‘how a trial on the merits would be conducted if a class were certified’ ”) (quoting BioPay LLC, 541 F.3d at 326-29).

Assuming that: (1) this case does go to trial, and (2) the experts are allowed to testify, and (3) a jury is asked for its verdict— that is, assuming a trial is not averted by rulings on the pending summary judgment motions (or a motion for directed verdict), and the experts are not excluded by rulings on the pending Daubert motions, and the parties do not agree to a bench trial — and further assuming the jury finds for plaintiffs, then there will have to be a determination of damages. This raises the practical problem of what precisely a jury will be asked regarding damages and how it might go about answering.

To oversimplify, the essence of plaintiffs’ ease will include two essential parts: (1) presentation of documents and witnesses to establish Sprint’s computerized commission-calculation system did not work properly, resulting in chronic underpayment to sales reps; and (2) explanation of its experts’ (BE’s) reconciliation process, ending with (i) a list of each class member and the damages he or she suffered, if any; and (ii) a grand total of underpayments.

Splint will respond primarily by: (1) presenting documents and witnesses to establish its computerized commission-calculation system was not as bad as plaintiffs say, even if it did have problems; (2) offering reasons why BE’s reconciliation model is wrong; and (3) presenting and explaining its own expert’s (Dr. Thornton’s) reconciliation process, ending with a competing list showing each class member suffered no damages. Regarding prongs 2 and 3, Sprint will attempt to show, for example, that BE made the following mistakes (among others) in its reconciliation model: “(i) inclusion of non-commissionable handset sales; (ii) failure to count deactiva-tions and exchanges; (iii) failure to distinguish between iDEN and CDMA devices (which was important for the 2005-2007 commission contracts); (iv) inclusion of Contract Renewals outside of the operative commission contracts; and (iii) [sic] failure to connect Add-ons/Contract Renewals to a RMS [in-store point-of-sale] transaction.” Sprint Reply at 9 (docket no. 632). Sprint will assert that, if these errors are corrected, an accurate reconciliation shows each plaintiff was actually overpaid. Ultimately, the case will largely boil down to which reconciliation model the jury believes is more correct.

What then should the jury be asked to do? The answer cannot be to enter onto a verdict form a damage amount for each of the 30,-000-plus class members. Nor can the answer be to give the jury only two choices — the number urged by plaintiffs (currently $102.2 million) or the number proffered by Sprint ($0). After all, the jury might conclude plaintiffs’ model is mostly right, but some of Sprint’s criticisms are valid, so the correct total damages number is (for example) $50 million. Or the jury might conclude plaintiffs’ model is not only entirely correct, but also that the model undercounts damages, and the final number should be (for example) $150 million. See Plaintiffs’ Response at 32 n.16 (docket no. 626) (“Plaintiffs anticipate their experts will testify they often took a conservative approach when addressing issues raised by Sprint’s faulty data, meaning the damages figure could be higher [than $102.2 million].”). How can the Court design the trial to allow the jury to exercise its discretion while also addressing the issue of individual damages — including plaintiffs’ admission that there are zero-damage plaintiffs?

One possibility is to employ the undersigned. See 4 Newberg on Class Actions § 12:5 (5th ed. June 2016 update) (“A court may appoint a special master to determine individual damages in a common fashion. A special master is usually used in complex cases that would otherwise be unmanageable due to individualized damages or other issues.”) (footnotes omitted). But it appears a quicker, cheaper, better method has been identified by Judge Lungstrum in his Urethane Antitrust MDL.

In that ease, plaintiffs claimed defendant Dow engaged in a five-year, horizontal price-fixing conspiracy for “polyether polyol,” a chemical ingredient used to manufacture polyurethane foam. In 2008, Judge Lungst-rum certified the purchaser-class, despite Dow’s protestations that damages were highly individualized, stating: “The fact that the damage element may involve more predominantly individualized issues does not defeat the fact that common issues will predominate the claim as a whole.” In re Urethane Antitrust Litig., 251 F.R.D. 629, 632 (D.Kan.2008), affirmed, 768 F.3d 1245 (10th Cir.2014).³⁵

Five years later, the case went to trial. Plaintiffs’ expert presented a complex economic model using multiple-regression statistical analysis and “opined that the class suffered damages ... in the amount of $496,680,486.” In re Urethane Antitrust Litig., 2013 WL 2097346 at *7 (D.Kan. May 15, 2013). The jury, however, awarded total damages in the lower amount of $400,049,039. Id. After the verdict, Dow filed a motion for judgment as a matter of law; one argument was “that the amount of damages was not sufficiently supported by evidence because the jury did not pick a damages figure specifically mentioned by [plaintiffs’ expert].” Id. at *7. Judge Lungstrum rejected this argument, noting “the jury was entitled to estimate damages” and “[t]he jury’s reduction ... to $400,049,039 could reasonably have been reached in a number of ways, as the jury could have accepted one or more of Dow’s arguments attacking that damage figure.” Id.

In addition, defendants filed a post-verdict motion to decertify the class; one argument was that, “under [the plaintiffs’ expert’s] model, a few class members did not suffer any damages, and ... each class member must suffer harm.” Id. at *1. Judge Lungst-rum rejected this argument too, because he “agree[d] with plaintiffs ... that all members of the class may be shown to have been impacted by a conspiracy that elevates prices above the competitive level, even if some members may have mitigated their damages or otherwise did not suffer damages that may be quantified. Moreover, Dow has not cited any authority supporting the argument that the presence of a few ‘zero-damages’ class members necessarily defeats certification.” Id.; see also id. at 6 (“the law does not support Dow’s argument that plaintiffs’ claim fails if they do not show injuries and damages suffered by each and every class member.”).

Having rejected Dow’s post-judgment motions, Judge Lungstrum then “granted the plaintiffs’ request to permit allocation of the award according to [them expert’s] damages model, with a pro rata reduction to reflect the jury’s award of a lesser amount.” In re Urethane Antitrust Litig., 768 F.3d 1245, 1252 (10th Cir.2014).

On appeal, the Tenth Circuit strongly endorsed Judge Lungstrum’s approach. First, Dow insisted “ it was entitled to show in individualized proceedings that certain class members could not have been injured by the alleged conspiracy.” Id. at 1254. The appellate court rejected this contention, concluding class certification was appropriate despite “[t]he presence of individualized damages issues.” Id. at 1255. Dow also reasserted that “the jury’s assessment was speculative because it deviated from [plaintiffs’ expert’s] figure and lacked any other evidentiary support.” Id. at 1268. The appellate court disagreed, stating “a jury can reduce an expert’s calculations on damages even when unable to ‘run the exact numbers and calculations of [a damages] model with mathematical certainty.’ ” Id. (quoting Medcom Holding v. Baxter Travenol Labs., Inc., 106 F.3d 1388, 1400-01 (7th Cir.1997)) (some internal quotation marks omitted).

Finally, the Tenth Circuit also affirmed Judge Lungstrum’s “decision to permit [pro rata] allocation of the damages award according to [plaintiffs’ expert’s] damages model.” Id. Dow sought reversal because the jury’s “reduction was based on a finding that certain class members suffered no injury,” and Judge Lungstrum’s approach left Dow “unable to have a jury determine which class members had suffered less damage than [the expert] had figured.” Id. at 1269 (emphasis added). But the appellate court “reject[ed] this argument because Dow has no interest in the method of distributing the aggregate damages award among the class members.” Id.

This Court can follow Judge Lungstrum’s approach in this case. The parties’ experts will present their competing commission-reconciliation models, and attack the accuracy of the other side’s. The jury will determine whether it believes Splint’s model or plaintiffs’ model is more accurate; and if it is plaintiffs’ model, the jury will further determine how accurate. As in the Urethane Antitrust MDL, the jury will be permitted to enter any total damages award that is supported by the evidence; if it is not the exact $102.2 million figure calculated by BE, then the award can be allocated pro rata. Of course, mathematical precision is not required — neither classwide nor individually. See Hoffer Oil Corp. v. Carpenter, 34 F.2d 589, 592 (10th Cir.1929) (stating that, in a breach of contract case, a “reasonable basis for computation, and the best evidence which is obtainable under the circumstances of the case and which will enable the jury to arrive at an approximate estimate of the loss, is sufficient.”); A to Z Rental, Inc. v. Wilson, 413 F.2d 899, 908 (10th Cir.1969) (“When the cause and existence of damages have been established [in a breach of contract case] with the requisite certainty, recovery will not be denied because the amount of such damages is difficult of ascertainment”). Notably, however, if the jury does award damages, the class members whom BE states were not underpaid will get no pro rata allocation, regardless.

In sum, by using the approach Judge Lungstrum adopted in his Urethane Antitrust MDL, as endorsed by the Tenth Circuit Court of Appeals, this Court will discharge concerns surrounding trial management and assessment of damages.³⁶

VII. Conclusion.

When this Court granted plaintiffs motion to certify, it concluded plaintiffs met the test for commonality because: (1) the question of “[w]hether defendants breached the contract by incorrectly determining commissions is an issue which undergirds every claim;” and (2) “All members of the proposed class base their claims on the same legal and factual theories — that defendants breached an agreement to pay commissions.” Certification Order at 15, 16 (docket no. 99). Moreover, the Court concluded plaintiffs met the test for predominance, because “[t]he record reveals a common nucleus of operative facts which are relevant to the dispute, and the common questions represent a significant aspect of the case which can be resolved for all members of the class in [a] single adjudication. * * * Specifically, the critical issues for trial are whether Sprint’s accounting methods systematically under reported commissions and interpretation of the plans and CAFs under Kansas law.” Id. at 20.

Contrary to the arguments Sprint advances in its motion to decertify, neither the fine details of this case as revealed by fact and expert discovery, nor close examination of the parties’ experts’ damages models undertaken by the Special Master and the Technical Advisor, supply good grounds for the Court to change its decision. Rather, the record shows there are numerous issues of fact and law shared by each plaintiff which are susceptible to resolution through common answers; the number and importance of those common issues outweigh other questions that are individual to each plaintiff, including the question of damages; and there exist mechanisms that will allow the Court to manage this litigation as a class through trial.

Furthermore, to the extent the Supreme Court modified the proper analysis of class certification after this Court issued its Certification Order, plaintiffs continue to meet all of the requirements of Fed. R. Civ. P. 23(a) and 23(b)(3). Accordingly, Sprint’s motion to decertify should be denied.

RESPECTFULLY SUBMITTED,

1 . The plaintiffs originally also asserted claims for: (3) quantum meruit; (4) promissory estoppel; and (5) unjust enrichment. The Court granted in part Sprint’s motion to dismiss, see docket no. 77, so plaintiffs now pursue only the KWPA and breach of contract claims, while reserving their rights on appeal.

2 . The Certification Order is published at Sibley v. Sprint Nextel Corp., 254 F.R.D. 662 (D.Kan.2008).

3 . See also id. at 3-4 ("The Advisor will supply neutral explanations in order to allow the Court to resolve the technical arguments and evidentia-ry issues within the parties’ Daubert motions and Sprint’s motion to decertify the class. The Advis- or will not create her own computer model or analysis to determine her own conclusion of how much Sprint owes the class (if anything); rather, her duties are limited to explaining the parties’ experts' existing analyses. The Advisor will examine and elucidate the existing expert reports, testimony, and calculations, as opposed to conducting an entirely independent damages calculation. Further, the Advisor will not supply legal analysis.’’).

4 . In their most recent expert reports, BE concluded Sprint underpaid the class by a total of $102.2 million, while Dr. Thornton concluded Sprint overpaid the class by a total of $75.0 million. See Plaintiffs’ Response at 18 (March 31, 2016) (docket no. 626) (citing BE Class Reconciliation Summary, docket no. 626-3, filed conventionally and not on ECF); Thornton’s Rebuttal Report at 59 n.106 (Jan. 28, 2016) (docket no. 618-26).

5 . To some extent, a Retail Consultant could track her own Gross Activations, because she would usually know how many customers to whom she sold a new calling plan. The Consultant could not readily track her own Net activations, however, because she would not usually have personal knowledge that a customer subsequently canceled his new service within 14 days. See docket no. 618-29 at 112 (2006 CAF further defining "New Net Activations” as "initial service activations that result in new subscriber IDs in the billing system which are not deactivated within 14 days of initial activation").

6 . Emblematic of how complicated Sprint made its compensation plans is the following single sentence that is a. partial definition of "Recurring Revenue”:

Recurring Revenue includes only revenue from the following items and services sold or activated under your sales ID during the measurement month: (i) Monthly Recurring Charges ("MRCs” — i.e„ service charges that are generally invoiced to customers on a recurring monthly basis) for Attachable or Add-on Services (i.e„ voice or data services such as Vision or Voice Command added on to a primary plan (a plan containing a certain predetermined number of minutes of services for a customer), as determined by the Company in its sole discretion) done on both New Net Activations (i.e. initial service activations that result in new subscriber IDs in the billing system which are not deactivated within 180 days of initial activation and are not intra-Company ports as defined in the New Activations section above) and for existing customers; (ii) Data revenue from select Bundled Service Plans (i.e., a primary service plan that contains both voice and data minutes&emdash;the dollar amount allocated to data is determined by the Company in its sole discretion, and the data portion is included as part of the Recurring Revenue Opportunity); and (iii) Modem Card Plans (primary plans on modem cards such as EVDO cards that are comprised of data minutes only).

Docket no. 618-29 at 5-6 (2006 CAF).

7 . In their briefs, plaintiffs set out this evidence in detail. See Daubert memo at 14-36 (docket no. 616); summary judgment memo at 14-46 (docket no. 638). Although Sprint insists it did not underpay its employees, it does not contest plaintiffs' assertion that the computer programs and processes it used to calculate commissions were often inaccurate and ineffective. See Sprint response brief on certification at 51-52 (docket no. 45) (characterizing the evidence highlighted by plaintiffs as "a wide variety of unanticipated computer issues that Sprint has already spent between $8,000,000 and $10,000,000 and 35,-000 - 40,000 hours to fix"). Indeed, the figures contained in the report of Sprint’s own expert, Dr. Thornton, reflect that nearly 100% of Sprint's contemporaneous monthly commission calculations for the class were wrong. See docket no, 638-1 at ¶4 (plaintiffs' analysis of Thornton's report shows that, "counting all over and underpaid results Dr. Thornton calculated [for] each [employee-pay] period, and dividing that figure by the total number of [employee-pay] periods Dr. Thornton calculated, [shows her overall conclusion] is [that] 99.97% of the class [was] paid incorrectly.”).

8 . Sprint earlier asserted Dr. Thornton's review "was not a commission reconciliation process," and instead was simply a search to see if there was "any kind of a systematic computer error.” Hearing tr. at 63 (March 3, 2014) (docket no. 626-19). In her most recent report, however, Dr. Thornton concedes she "conducted] a transaction-by-transaction review for each individual.” Sprint Reply at 3 n.4 (docket no. 632) (citing Thornton's 9/28/15 report at 4 n.6 (docket no. 616-87)). Indeed, Dr. Thornton believes her reconciliation is "more accurate than Sprint's” original commission calculations. Depo. at 139 (docket no. 626-18).

9 . Another reason the parties reach different conclusions is because, in some instances, Sprint's written compensation plans were ambiguous, and the experts’ interpretations are different. For example, the 2006 CAF confers commission based partly on the number of new phones sold, but the CAF is silent regarding whether a “free” phone included with the sale of a calling plan is counted. The parties’ experts treated these "free” phones differently — plaintiffs counted them and Sprint did not. Most of the difference in the parties' calculations, however, relates to how they used and mined the existing data, not on contract interpretation.

10 . Cf. Day v. Celadon Trucking Servs., Inc., 827 F.3d 817, 832 (8th Cir.2016) (concluding it is appropriate to ”requir[e] the movant to bear the burden of showing that the district court mistakenly maintained class certification”); id. at n. 5 (canvassing different rulings on the issue of who bears the burden on a decertification motion).

11 . Harlow is virtually identical to this case — it shares the same counsel for both sides, the same experts, the same basic facts, and the same claims — except it was brought by employees working in "Sprint's Business Direct Channel” instead of employees working in Sprint’s retail stores. Harlow, 254 F.R.D. at 419, The Honorable John W. Lungstrum concluded certification in Harlow was appropriate under Rule 23(b)(3), lending independent support to the Court’s original certification decision in the instant case. See-id. at 420-25 (finding plaintiffs met their "strict burden of proof” of showing "numerosity, commonality, typicality, and adequacy of representation,” as well as predominance and superiority). After Judge Lungstrum certified the class in Harlow, he transferred the case to the Honorable Kathryn Vratil, "because th[e] case is so similar to the earlier-filed Sibley v. Sprint Nextel Corp." Id. at 425,.

12 . Sprint writes: "Since the entry of the Court’s 2008 Certification Order, the Supreme Court issued its decisions [sic] in Dukes, materially changing the analysis of the standards governing class certification under Rule 23, particularly in regard to the 'commonality’ requirement of Rule 23(a).” Decertification Memo at 12 (emphasis added) (docket no. 620). When Dukes was new law, this Court tentatively agreed that it changed the Rule 23 analysis. See In re Motor Fuel Temperature Sales Practices Litig., 279 F.R.D. 598, 607 (D.Kan.2012) (Vratil, J.) (“Dukes ... arguably heightened the commonality requirement under Rule 23(a) and narrowed the permissible scope of a Rule 23(b)(2) class”). But more recent Circuit Court opinions hold that Dukes does not constitute a change (material or otherwise) in class certification standards. See Brown v. Nucor Corp., 785 F.3d 895, 920 (4th Cir.2015) (Dulces "did not change, nor purport to change, the Rule 23(b)(3) analysis”); Rodriguez v. National City Bank, 726 F.3d 372, 381 n. 8 (3rd Cir.2013) ("Although the Supreme Court’s decision in Dukes is an intervening and pointedly clear explication of the law, it did not announce any change in the test for determining commonality.”); Schell v. OXY USA Inc., 2013 WL 4857686 at *5 (D.Kan. Sept. 11, 2013) (Marten, J.) (denying decertification based on “the Tenth Circuit's ruling in Roderick v. XTO Energy, Inc.[, 725 F.3d 1213 (10th Cir.2013) ] that Dulces did not change the standard”).

13 . Sprint suggests Comcast is new law that undermines the predominance analysis the Court set out in its Certification Order. Other courts disagree. See Brown v. Electrolux Home Prods., Inc., 817 F.3d 1225, 1238 (11th Cir.2016) ("Com-cast did not change the law about the effect of individual damages on predominance.”); id. at 1239 ("other courts agree that Comcast did not change the law that a class action can sometimes be maintained notwithstanding the need to prove individual damages”) (citing cases).

14 . Under the applicable law of Kansas, damages is an essential element of a claim for breach of contract. City of Andover v. Southwestern Bell Tel., L.P., 37 Kan.App.2d 358, 153 P.3d 561, 565 (2007). To prevail on a derivative KWPA claim, the plaintiff must prove his employer failed to pay him "all wages due,” K.S.A. § 44-314(a), which is the same showing.

15 .See Plaintiffs' Response at 18 (March 31, 2016) (docket no. 626) (citing BE Class Reconciliation Summary, docket no. 626-3, filed conventionally and not on ECF). BE's most recent analysis concludes that, of 32,941 employee class-members, 3,498 (10.6%) were overpaid, 455 (1.4%) were correctly paid, and 28,988 (88.0%) were underpaid.

16 . Plaintiffs do argue Sprint's alleged failure to pay commissions to the class was "willful,” and that double damages are therefore appropriate under Kan. Stat. Ann. § 44-315. Amended complaint at ¶¶ 44-45 (docket no. 8). But this issue of "intent” is not individual to each class-member.

17 . Sprint is quick to point out that determination of whether and how much commission an employee earned sometimes "involv[ed] human discretion, such as evaluations of quotas and assessments of appeals.” Decertification Memo at 5 (docket no. 620). The overwhelming majority of Sprint’s commission calculations, however, were purely automated.

18 . Overarching problems identified by plaintiffs’ experts include: (1) "incomplete and poor quality data,” (2) “lack[] of any meaningful centralized data or system management” — specifically " 'si-loing,' which is a term used to describe when a company manages data or systems vertically rather than across all systems,” and (3) ”fail[ure] to standardize[ ] and synchronize data across systems.” BE’s expert report at 1-2 (docket no. 618-24), General examples of how these problems manifested include: (a) "assignment of multiple and duplicate employee identifiers to employees," (b) "[calculation] errors or omissions in Sprint’s compensation plans,” (c) deactivation data that "does not appear to be accurate or complete,” (d) "Compensation Statements that reflect certain employees were working in certain job positions during periods when the commissions plans do not contain those job positions,” (e) use of "SKU lists [that] did not contain all the SKUs in effect,” (f) “missing or conflicting data ... concerning SOC pricing used in commissions,” and so on. Id. at 11, 16, 23, 24, 27, 33. A minute example of a specific problem is that Sprint's "data showed SOC SMS-FPPT, a family power pack text plan, had 13 different MRC [monthly recurring charge] values ranging from $2.50 to $99.99 in period 179, with the most frequent MRC value being $20.00.” Id. at 38. Sprint's characterization of plaintiffs’ proofs as simply "anecdotal evidence that Sprint’s commissions systems were difficult to administer, or that they were disliked by some, or that some individuals found the procedures cumbersome,” is steep downplay. Sprint Reply at 19 n.10 (docket no. 632).

In any event, plaintiffs are not required to match each alleged commission underpayment to a specific reason it occurred. They only need show they were, in fact, underpaid, and by how much. Proof of commissions earned and payments received accomplishes this showing.

19 . Sprint cites several other cases to support its assertion that "the commission contracts themselves cannot serve as the 'glue' to hold Plaintiffs' claims together.” Decertification Memo at 14 (docket no. 620). But in none of these cases were all of the class members affected in a uniform way by identical policies or contracts, with the effect measurable by computer program, as they are in this case. See Tabor v. Hilti, Inc., 703 F.3d 1206, 1228-30 (10th Cir.2013) (denying certification to female employees in a failure-to-promote case, because the crux of the claim was "a highly discretionary policy for granting promotions”); In re Bank of America Wage & Hour Empl. Litig., 286 F.R.D. 572, 588 (D.Kan.2012) (Lungstrum, J.) (plaintiffs did not even allege there was a "systemic failure to pay overtime to employees,” nor that "all class members ... are affected by the employer’s allegedly unlawful policy;” thus, each plaintiffs' claims for overtime required "individualized assessments”); Burdette v. Vigindustries, Inc., 2012 WL 405621 at *7 (D.Kan. Feb. 8, 2012) (Robinson, J.) (in a negligence and nuisance case stemming from sinkholes allegedly caused by defendant, each plaintiff’s claim was "based on the homeowner's use and enjoyment of their property," which necessarily required "the individual Plaintiff’s testimony").

Cf. In re Checking Account Overdraft Litig., 281 F.R.D. 667, 677 (S.D.Fla.2012) (certifying a class of bank customers and finding predominance because "the agreements at issue here are uniform form contracts offered on a take-it-or-leave it basis and were not the product of any individual negotiation. * * * It is the form contract, executed under like conditions by all class members, that best facilitates class treatment.”) (internal quotation marks and citation omitted); In re Universal Service Fund Telephone Billing Practices Litig., 219 F.R.D. 661, 667, 677 (D.Kan.2004) (Lungstrum, J.) (certifying a class of customers of Sprint and other wireless carriers who claimed breach of "provisions in [their] standard form contracts,” and noting "the fact that individual proof as to the amount of damages may be necessary does not preclude class certification”) (citation omitted).

20 . Sprint misreads "systematic” to mean either "single,” or "having the same effect on every class member." As reflected above, neither the plaintiffs nor the Court ever used the term "systematic” to mean "single.” Further, the Court understood correctly that a given problem with Sprint's commission-calculation system might or might not have the same effect on every member of the class — that is, a given problem might "affect a large number of employees ... [or] a relatively small number of employees.” Certification Order at 7-8 (docket no. 99). The system as a whole, however, was designed to and did have a global and common impact on all Sprint sales rep employees — it calculated their commission pay using universally-applied equations and comprehensive, company-wide data. Thus, the alleged problems were both systemic and systematic. Plaintiffs’ theory was clear when they filed their original motion for class certification and it remains clear, and unchanged, today.

21 . The Special Master makes this statement without reference or prejudice to any arguments the parties made in their cross-motions for summary judgment or Daubert motions.

22 . See plaintiffs’ motion for summary judgment at 70-78 (docket no. 638) (asserting seven different reasons that Sprint’s argument "that the Plan prohibits Plaintiffs from bringing a lawsuit unless they prove they first exhausted Sprint’s internal reporting procedure for disputed commissions” is "contrary to the plain language of the Plan and violates the law”). In other words, Sprint’s affirmative defense presents a common question of law.

23 .The Tabor court concluded the defendant’s "policy of allowing broad discretion to local managers ‘is just the opposite of a uniform employment practice that would provide the commonality needed for a class action.’ ” Id. at 1228-29 (quoting Dukes, 564 U.S. at 365, 131 S.Ct. 2541). Sprint’s computerized commission-calculation system is certainly not the "opposite of a uniform employment practice.”

24 . As noted earlier, while the concerns Roderick raises are important, Roderick is distinguishable on its facts. The Roderick plaintiffs' lease contracts were not uniform, having at least "twenty different categories of royalty provisions." Id. at 1216. In this case a single, uniform contract applied to every plaintiff in this case at any given time. Further, the "uniform payment methodology” that the Roderick district court identified as the common question was actually not common, given lease-by-lease differences in “gas marketability.” Id. at 1219. In this case, the payment methodology is homogenous.

25 . As one court observed, in light of "the fact that proving actual injury if suits were brought individually would still require the same types of assumptions made by [the plaintiff’s expert] in his report” regarding class-wide damages, "common questions predominate in this action.” Gutierrez v. Wells Fargo & Co., 2010 WL 1233810 at *14 (N.D.Cal. Mar. 26, 2010) (emphasis added). Further, this fact supported the conclusion that dass treatment would not “truncate the required substantive elements of proof by each claimant in violation of the Rules Enabling Act.” Id.

26 . Again, the Special Master makes this statement without reference or prejudice to any arguments the parties made in their cross-motions for summary judgment or Daubert motions.

27 . Then-Judge Sotomayor further noted that “[t]here are a number of management tools available to a district court to address any individualized damages issues that might arise in a class action, including: (1) bifurcating liability and damage trials with the same or different juries; (2) appointing a magistrate judge or special master to preside over individual damages proceedings; (3) decertifying the class after the liability trial and providing notice to class members concerning how they may proceed to prove damages; (4) creating subclasses; or (5) altering or amending the class." In re Visa, 280 F.3d at 141 (footnote omitted).

28 . In the Supreme Court's most recent class action case, Tyson Foods, Inc. v. Bouaphakeo, — U.S. —, 136 S.Ct. 1036, 194 L.Ed.2d 124 (2016), the Court approved plaintiffs’ use of a "fancy model” to establish liability and measure damages. Specifically, the Court held that representative proof from a statistical sample of the class could be used to show predominance of common questions of law or fact. Here, the model offered by plaintiffs' experts does not rely upon a statistical sample — it goes even further and calculates the amount of commissions allegedly underpaid for each individual class member.

On a related note, shortly before the Special Master finalized this Report, Sprint offered "supplemental authority” that touches on Tyson Foods. See docket no. 662 (discussing Senne v. Kansas City Royals Baseball Corp., 2016 WL 3940761 (N.D.Cal. July 21, 2016)). It suffices to say the facts of Senne are easily distinguishable from this case. See, e.g., id. at *86 ("Plaintiffs [minor league baseball players] assert claims under the laws of eight states;” in this case, only Kansas law applies); id. at *77 (plaintiffs “negotiate[d] with [22 different MLB] Clubs over the terms of their compensation;” in this case, plaintiffs have adhesion contracts with one employer); id. at *80 (the court would have to determine if "the types of activities in which plaintiffs' engaged ... were found to constitute 'work;' ” in this case, commissionable activity is extremely carefully defined); id. at *87 (plaintiffs sought "to prove their claims with common proof by using representative data, namely, a survey;” in this case, plaintiffs' proof of liability and damages is not made via extrapolation, it is made for every single class member through use of Sprint’s entire dataset).

29 . Although more than 25 years old, the facts of Smith are strikingly similar to those in this case. Like Sprint, Smith defendant MCI provided wireless telecommunications products and services, and it paid commissions to its sales reps. Also, like Sprint, MCI; (1) used "various computer systems” to track information necessary to calculate commissions; (2) produced, during discovery, "in-house memoranda indicating] that [there were] difficulties” with its “Integrated Commission System (ICS)," such as inaccurate calculations and failure to recognize sales; and (3) was sued by plaintiffs for (among other claims) breach of contract and violation of the Kansas Wage Payment Act, "based on MCI’s failure to pay salesperson's commissions.” Smith, 124 F.R.D. at 670-71.

MCI's principal argument against class certification could have been written by Sprint: "determining damages will involve the painstaking process of comparing accounts sold and commissions paid during the time each of the different commission plans and addenda [were] in effect.” Id. at 677 (internal quotation marks omitted). The court rejected this argument, concluding;

Smith’s complaint alleges that because of MCI's actions, salespersons were cheated out of the commissions to which they were entitled. Thus, the members of the proposed class all suffered the same type of damages. Further, the fact that the proposed class members worked under different compensation plans does not indicate that the damages question is predominately individualized. Of course, the amount of damages allegedly suffered by members of the prospective class will vary depending on the plans under which they worked and the number of sales on which they were denied commissions. However, the actual amount of damages is invariably an individual question and does not defeat class action treatment. Thus, if the computation of damages following a ruling in favor of the class is a largely mechanical task, then the existence of individualized claims for damages seems to offer no barrier to class certification.

Id. at 677 (internal citations and quotation marks omitted).

The actual computation of commissions in Smith was far simpler than it is in this case — the plaintiff quaintly suggested he could "use a pocket calculator on the actual customer bills to determine the amount owed in commissions,” id.— but it was the ultimately the fact that "calculation of damages following a ruling in favor of the class would be a largely mechanical task” that led the court to conclude the commission calculation "is not an individualized process which predominates over the common issues presented.” Id. at 678 (emphasis in original).

Yet another faetually-similar case is Heartland Communic’ns, Inc. v. Sprint Corp., 161 F.R.D. 111 (D.Kan. 1995) (Lungstrum, J.), where Sprint's outside sales reps sued for breach of commission contracts. It suffices to say the court certified the class in that case, too.

30 . See City of Andover v. Southwestern Bell Tel., L.P., 37 Kan.App.2d 358, 153 P.3d 561, 565 (2007) ("The essential elements of an action based on a contract are: (1) The existence of a contract between the parties; (2) Sufficient consideration to support the contract; (3) The plaintiffs performance or willingness to perform in compliance with the contract; (4) The defendant's breach of the contract; and (5) Damages to plaintiff caused by the breach.”) (quoting Pattern Jury Instructions of Kansas Civ.3d 124.01-A).

31 .Sprint notes plaintiffs have identified 12% of the class as "zero-damage” plaintiffs, which is a substantial fraction, not just "a few." See footnote 15, above. Plaintiffs observe, however, that their calculations show 95.6% of the class was underpaid at least one month; even Sprint calculates that 58% of the class was underpaid at least one month. See Plaintiffs' Response at 18 (docket no. 626) (plaintiffs' chart showing percentages of over- and underpaid employees); see Thornton’s rebuttal report at 59 n.106 (docket no. 616-90) (Technical Advisor’s analysis of Dr. Thornton's data-set counts2015update.sas7bdat). Plaintiffs imply that Sprint thus breached its contractual duty to pay the required monthly commission to virtually every class member, not just 88% of them, but that damages flowing from those monthly breaches are sometimes overwhelmed by set-off, in the form of overpayments during other months. Compare Sprint’s answer at 20 (docket no. 17) (asserting the defense that "Defendants have the right of recoupment and/or setoff against Plaintiffs and each putative class member who received compensation they did not earn and were not otherwise entitled to receive.”). This is an analytically-different view of breach and damages, and provides a lower measure of “zero-damage” plaintiffs (as opposed to plaintiffs who suffered above-zero damages but also countervailing offsets). The parties do not fully engage with the question of which view is more correct. In any event, the presence of zero-damage plaintiffs does not alter the Special Master's conclusion on the propriety of class certification.

32 . Sprint also seeks to have it both ways when, after complaining that plaintiffs have identified class members who suffered zero-damages, it suggests “there is the possibility that numerous plaintiffs have the potential to recover commissions even when there is none due under the terms of the commission contracts.” Sprint Reply at 7 (docket no. 632). If Sprint wins at trial, no plaintiff will recover commissions when none are due; if plaintiffs win, the jury will have accepted plaintiffs' damages model and its individualized determinations, including identification of the "zero-damages” class members — so, (again) no plaintiff will recover commissions when none are due.

33 . Sprint refers to superiority only once, in passing, in each of its two briefs. Decertification Memo at 21 (docket no. 620); Sprint Reply at 10 (docket no. 632)

34 . The undersigned served as Special Master in the Welding Fumes MDL. The Court had similar concerns in that case regarding dueling experts' presentations of the cause, location, pattern, and symptoms of cellular degeneration in the brain. Post-verdict interviews with jurors following the eight bellwether trials, however, repeatedly revealed their capacity to decipher, and reach reasoned conclusions based upon, obscure and sophisticated neuropathological evidence.

35 . See also id. at 639 ("The court is not nearly as persuaded that the issue of damages is as amenable to class-wide proof as the issues of antitrust conspiracy and impact in light of the myriad of products, pricing structures, individual negotiations, and contracts at issue. * * * The court believes that the most likely scenario is that plaintiffs will be able to use a formulaic approach to damages through [expert] testimony with respect to some damage calculations, but others may require individualized determinations. The possibility that individual issues may predominate the issue of damages, however, does not defeat class certification by making this aspect of the case unmanageable.”).

36 . Newberg lists several "methods for determining individual damages." 4 Newberg on Class Actions § 12:5 (5th ed. June 2016 update). In addition to using special masters, courts can allow use of calculation models: "[formulae for calculating class aggregate damages may also be based on records, typically the defendant’s records. * * * Even in instances in which the defendant’s records cannot lead to exact calculations, they often form a rubric or data point that a formula can use.”). Id. Newberg concludes: "Any of these methods — or combinations of them — help enable courts to resolve individual damages issues in an efficient enough manner that such issues will not predominate over the common issues in the case; in other words, these approaches enable certification of money-damage class suits.” Id. (emphasis added). If necessary, then, the Court can create a hybrid approach combining Judge Lungstrum's method, a special master, and any other appropriate technique.